The goal of this project is to use proteomics to investigate a post-translational modification that is associated with Chromogranin A, the recently discovered autoantigen for the diabetogenic CD4 T cell clone BDC-2.5. Preliminary data shows that the antigenicity of ChgA-derived peptides can be increased significantly upon treatment with the enzyme tissue transglutaminase (TGase). Additional data indicates that natural antigen obtained from ?-cell tumors contains an aldehyde or keto functional group. In specific aim 1 we will investigate the chemical nature of the TGase-treated peptide. For this we will test the effect of TGase treatment on altered ChgA peptides, and use proteomic purification, mass spectrometric and biochemical derivatization methods to investigate the role of peptide TGase interactions. In specific aim 2 we will investigate the presence of an aldehyde or keto group in antigen obtained from pancreatic ?-cells. For this we will use biochemical derivatization and mass spectrometric methods to purify derivatized antigen and to identify the chemical nature of the modification. TGase-treated peptides and proteins containing an aldehyde or keto group can form covalent bonds with primary amines, such as lysines. The major histocompatibility complex class II (MHC II) molecule I-Ag7 contains several lysine residues and in specific aim 3 we will investigate whether TGase-treated peptides or ?-cell antigens form a covalent bond with I-Ag7. For this we will covalently label I-Ag7 using ?-cell antigen and biotin-labeled peptides that were treated with TGase. Proteomic and mass spectrometric methods will be used to purify and identify the covalent attachment. Post-translationally modified antigens in type 1 diabetes (T1D) have so far not been described and could play be a key role in the initiation of this disease.